Three-Dimensional Extended Object Tracking and Shape Learning Using Gaussian Processes

In this article, we investigate the problem of tracking objects with unknown shapes using 3-D point cloud data. We propose a Gaussian process-based model to jointly estimate object kinematics, including position, orientation, and velocities, together with the shape of the object for online and offline applications. We describe the unknown shape by a radial function in 3-D, and induce a correlation structure via a Gaussian process. Furthermore, we propose an efficient algorithm to reduce the computational complexity of working with 3-D data. This is accomplished by casting the tracking problem into projection planes, which are attached to the object's local frame. The resulting algorithms can process 3-D point cloud data and accomplish tracking of a dynamic object. Furthermore, they provide analytical expressions for the representation of the object shape in 3-D, together with confidence intervals. The confidence intervals, which quantify the uncertainty in the shape estimate, can later be used for solving the gating and association problems inherent in object tracking. The performance of the methods is demonstrated both on simulated and real data. The results are compared with an existing random matrix model, which is commonly used for extended object tracking in the literature.

Automated summary

The article introduces a Gaussian process-based model for tracking objects with unknown shapes, estimating object kinematics and shape, and handling 3-D point cloud data. The algorithm reduces computational complexity by projecting to planes and provides confidence intervals for shape estimation uncertainty, demonstrating excellent performance for object tracking.